[PATCH] D18566: [x86] use SSE/AVX ops for non-zero memsets (PR27100)

Joerg Sonnenberger via llvm-commits llvm-commits at lists.llvm.org
Tue Mar 29 11:56:34 PDT 2016


On Tue, Mar 29, 2016 at 06:50:42PM +0000, Sanjay Patel via llvm-commits wrote:
> But as the test cases show, there's some ugliness here:
> 1. The i386 (Windows) test expands to use 32 stores instead of a 'rep stosl'. Is that better or worse? (I'm not sure why this change even happens yet.)

Hard to say given the overhead of rep, but it seems excessive.

> 2. The memset-2.ll tests look quite awkward in the way they splat the byte value into an XMM reg; imul isn't generally cheap.

It isn't?

Joerg


More information about the llvm-commits mailing list